Study on D-Call Net Effectiveness by using ITARDA Macro Database with the Emergency Transport Database (2nd Report)

Toru Kiuchi (ITARDA)･Nobuo Saito (Japan Mayday Service)･Masayuki Shirakawa (ITARDA)

At the JSAE spring session five years ago, the authors attempted to analyze the effectiveness using the 2015 and 2016 matched data, but did not come close to confirming a clear effect. Since the number of vehicles equipped with AACN systems has expanded significantly in recent years, we attempted to analyze the effectiveness by matching emergency transport data with automatic reporting data for the most recent period from 2020 to 2022.

Threshold for Activating the Doctor Dispatch System in the Advanced Automatic Collision Notification (D-Call Net) Algorithm Ver.2017

Tomokazu Motomura (Hokusoh HEMS Nippon Medical School Chiba Hokusoh Hospital/Nippon Medical School/D-Call Net Study Group)･Tetsuya Nishimoto (D-Call Net Study Group/Nihon University)･Hirotoshi Ishikawa (D-Call Net Study Group/HEM-Net)･Kazuki Mashiko (Hokusoh HEMS Nippon Medical School Chiba Hokusoh Hospital/Nippon Medical School)･Kunihiro Mashiko (D-Call Net Study Group/HEM-Net/Minami Tama Hospital)･Yoshiaki Hara (Hokusoh HEMS Nippon Medical School Chiba Hokusoh Hospital/Nippon Medical School)･Nobuya Kitamura (Kimitsu Chuo Hospital/Japanese Society for Aeromedical Services, Doctor Helicopter Committee)

The threshold value for algorithm Ver. 2017, which activates a doctor dispatch system such as Doctor-heli with automatic accident notification, was investigated. The threshold value for reducing the under-triage rate to less than 10% was determined for 326 traffic accident investigation cases that were transported to nms Chiba Hokusoh Hospital. [1] The age of the crew member is known: 10% [2] The age of the crew member is unknown: 20%.

Quantifying the Relationship between Emergency Transport Time and Injury Severity to Improve Survival Rates
-Analysis of Australian Emergency Transport Case Data-

Kazuhiro Kubota･Tetsuya Nishimoto (Nihon University)･Giulio Ponte (University of Adelaide)

The relationship between transport time and injury severity was quantified by analyzing data from emergency transport cases in Australia. The rate of minor injuries decreases as transport time increases, with the minor injury rate being less than 60% at 120 minutes. Transport time from the scene of a crash varies depending on the means of transportation, such as helicopters or ambulances. The relationship between transport time and the fatality and serious injury rate can be used to select the optimal means of transport based on distance and the availability of medical resources.

Accident and Injury Prediction Maps in Vehicle-to-Vehicle Collision Based on Accident and Road Information using Deep Learning

Yusuke Miyazaki･Tsubasa Miyazaki (Tokyo Institute of Technology)･Koji Kitamura (AIST)･Fusako Sato (JARI)

Conventional accident prediction models have been constructed using only structured data expressed in tabular form. However, it is difficult to incorporate environmental information such as intersection geometry into structured data alone. Therefore, this study developed a multimodal deep learning model that combines structured data and road image data to construct a vehicle accident prediction model.